arm_cmplx_mult_cmplx_q31.c

00001 /* ----------------------------------------------------------------------  
00002 * Copyright (C) 2010 ARM Limited. All rights reserved.  
00003 *  
00004 * $Date:        29. November 2010  
00005 * $Revision:    V1.0.3  
00006 *  
00007 * Project:      CMSIS DSP Library  
00008 * Title:        arm_cmplx_mult_cmplx_q31.c  
00009 *  
00010 * Description:  Q31 complex-by-complex multiplication  
00011 *  
00012 * Target Processor: Cortex-M4/Cortex-M3
00013 *  
00014 * Version 1.0.3 2010/11/29 
00015 *    Re-organized the CMSIS folders and updated documentation.  
00016 *   
00017 * Version 1.0.2 2010/11/11  
00018 *    Documentation updated.   
00019 *  
00020 * Version 1.0.1 2010/10/05   
00021 *    Production release and review comments incorporated.  
00022 *  
00023 * Version 1.0.0 2010/09/20   
00024 *    Production release and review comments incorporated.  
00025 * -------------------------------------------------------------------- */ 
00026  
00027 #include "arm_math.h" 
00028  
00029 /**  
00030  * @ingroup groupCmplxMath  
00031  */ 
00032  
00033 /**  
00034  * @addtogroup CmplxByCmplxMult  
00035  * @{  
00036  */ 
00037  
00038  
00039 /**  
00040  * @brief  Q31 complex-by-complex multiplication  
00041  * @param[in]  *pSrcA points to the first input vector  
00042  * @param[in]  *pSrcB points to the second input vector  
00043  * @param[out]  *pDst  points to the output vector  
00044  * @param[in]  numSamples number of complex samples in each vector  
00045  * @return none.  
00046  *  
00047  * <b>Scaling and Overflow Behavior:</b>  
00048  * \par  
00049  * The function implements 1.31 by 1.31 multiplications and finally output is converted into 3.29 format.  
00050  * Input down scaling is not required.  
00051  */ 
00052  
00053 void arm_cmplx_mult_cmplx_q31( 
00054   q31_t * pSrcA, 
00055   q31_t * pSrcB, 
00056   q31_t * pDst, 
00057   uint32_t numSamples) 
00058 { 
00059   q31_t a, b, c, d;                              /* Temporary variables to store real and imaginary values */ 
00060   uint32_t blkCnt;                               /* loop counters */ 
00061  
00062   /* loop Unrolling */ 
00063   blkCnt = numSamples >> 2u; 
00064  
00065   /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.  
00066    ** a second loop below computes the remaining 1 to 3 samples. */ 
00067   while(blkCnt > 0u) 
00068   { 
00069     /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1].  */ 
00070     /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i].  */ 
00071     a = *pSrcA++; 
00072     b = *pSrcA++; 
00073     c = *pSrcB++; 
00074     d = *pSrcB++; 
00075  
00076     /* store the real result in 3.29 format in the destination buffer. */ 
00077     *pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33)); 
00078     /* store the imag result in 3.29 format in the destination buffer. */ 
00079     *pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33)); 
00080  
00081     a = *pSrcA++; 
00082     b = *pSrcA++; 
00083     c = *pSrcB++; 
00084     d = *pSrcB++; 
00085  
00086     /* store the result in 3.29 format in the destination buffer. */ 
00087     *pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33)); 
00088     /* store the result in 3.29 format in the destination buffer. */ 
00089     *pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33)); 
00090  
00091     a = *pSrcA++; 
00092     b = *pSrcA++; 
00093     c = *pSrcB++; 
00094     d = *pSrcB++; 
00095  
00096     /* store the result in 3.29 format in the destination buffer. */ 
00097     *pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33)); 
00098     /* store the result in 3.29 format in the destination buffer. */ 
00099     *pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33)); 
00100  
00101     a = *pSrcA++; 
00102     b = *pSrcA++; 
00103     c = *pSrcB++; 
00104     d = *pSrcB++; 
00105  
00106     /* store the result in 3.29 format in the destination buffer. */ 
00107     *pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33)); 
00108     /* store the result in 3.29 format in the destination buffer. */ 
00109     *pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33)); 
00110  
00111     /* Decrement the blockSize loop counter */ 
00112     blkCnt--; 
00113   } 
00114  
00115   /* If the blockSize is not a multiple of 4, compute any remaining output samples here.  
00116    ** No loop unrolling is used. */ 
00117   blkCnt = numSamples % 0x4u; 
00118  
00119   while(blkCnt > 0u) 
00120   { 
00121     /* C[2 * i] = A[2 * i] * B[2 * i] - A[2 * i + 1] * B[2 * i + 1].  */ 
00122     /* C[2 * i + 1] = A[2 * i] * B[2 * i + 1] + A[2 * i + 1] * B[2 * i].  */ 
00123     a = *pSrcA++; 
00124     b = *pSrcA++; 
00125     c = *pSrcB++; 
00126     d = *pSrcB++; 
00127  
00128     /* store the result in 3.29 format in the destination buffer. */ 
00129     *pDst++ = (q31_t) ((((q63_t) a * c) >> 33) - (((q63_t) b * d) >> 33)); 
00130     /* store the result in 3.29 format in the destination buffer. */ 
00131     *pDst++ = (q31_t) ((((q63_t) a * d) >> 33) + (((q63_t) b * c) >> 33)); 
00132  
00133     /* Decrement the blockSize loop counter */ 
00134     blkCnt--; 
00135   } 
00136 } 
00137  
00138 /**  
00139  * @} end of CmplxByCmplxMult group  
00140  */